- An AI-enabled digital stethoscope can be a simple and effective way to diagnose cardiac dysfunction during and soon after pregnancy.
- The stethoscope records electrocardiograms and heart sounds and then uses an AI algorithm to predict the likelihood of having a weakened heart muscle.
- AI-guided screening using the digital stethoscope identified twice as many cases of pregnancy-related cardiomyopathy among women in Nigeria compared to standard clinical care.
Embargo until 9:45 a.m. ET, Monday, November 13, 2023
PHILADELPHIA, November 13, 2023 — Electrocardiogram (ECG)-based screening using an artificial intelligence-enabled digital stethoscope can detect peripartum cardiomyopathy, a disease of the heart muscle in pregnant and newly pregnant women, at twice the rate of standard obstetric care , including clinical EKGs, according to late-breaking science presented today at the American Heart Association’s Scientific Sessions 2023. The meeting, 11.-13. November in Philadelphia, is a leading global exchange of the latest scientific advances, research and evidence-based clinical practice updates in cardiovascular science.
In a study of nearly 1,200 Nigerian women who were pregnant or had recently given birth, peripartum cardiomyopathy was detected twice as often among study participants when ECG testing was performed with a digital stethoscope using an AI algorithm, compared to clinical ECG in addition to routine obstetric care.
“We demonstrated for the first time in an obstetric population that AI-guided screening using a digital stethoscope improved the diagnosis of this potentially life-threatening and treatable condition,” said study lead author Demilade A. Adedinsewo, MD, MPH, an assistant professor of Medicine in the Division of Cardiovascular Medicine at the Mayo Clinic in Jacksonville, Florida.
“This research may change current clinical practice from one that is reactive and symptom-driven to a more proactive approach to identifying pregnancy-related cardiac dysfunction using a simple, inexpensive and effective screening tool,” she said. “Earlier diagnosis would facilitate prompt and appropriate treatment of cardiomyopathy and reduce associated morbidity and mortality.”
Peripartum cardiomyopathy is a type of heart failure that can affect women late in pregnancy or after pregnancy. The disease weakens the heart, resulting in a decrease in the amount of blood pumped from the heart to other parts of the body. It is typically diagnosed toward the end of pregnancy or in the months after and can be difficult for health care professionals to detect because many of the symptoms are similar to those seen in normal pregnancy, such as shortness of breath and swelling in the feet and legs, according to the American Heart Association.
The rate of peripartum cardiomyopathy is somewhat low in the United States, affecting 1 in every 1,000 to 4,000 pregnancies, according to the association. It is more prevalent in Nigeria, which has the highest reported incidence of peripartum cardiomyopathy worldwide, affecting as many as 1 in 96 pregnancies, according to Adedinsewo.
The AI screening included a digital stethoscope to record the heart’s electrical activity and record heart sounds. The women in the AI intervention group also had a study-ordered echocardiogram at study enrollment to validate the effectiveness of the AI algorithm in this group. The AI-enabled stethoscope used an algorithm originally developed with 12-lead ECG data and modified for use with a single-lead ECG recorded with a digital stethoscope, enabling it to predict the likelihood of left ventricular dysfunction. Participants in the control group had traditional EKG testing without the use of the AI-assisted digital stethoscope, and their results were evaluated by healthcare professionals.
An echocardiogram was used to measure left ventricular ejection fraction, which measures the heart’s pumping ability. A normal ejection fraction level for the heart is between 55% and 70%. In this study, a left ventricular ejection fraction of less than 50% was the criteria for a cardiomyopathy diagnosis.
Between August 2022 and September 2023, the study enrolled more than 1,000 Nigerian women who were pregnant or had given birth to a baby in the previous 12 months.
The study found:
- Pregnancy-related cardiomyopathy was detected in 4% of pregnant and postpartum women in the group screened with the AI-enabled digital stethoscope.
- The detection rate of cardiomyopathy was 1.8% in the control group, suggesting that half of the cases of cardiomyopathy are likely to be undetected with usual care.
“Although we expected AI-guided screening to improve the diagnosis of cardiomyopathy, we did not expect the rate of cardiomyopathy diagnosis to double,” Adedinsewo said. “Further large trials enrolling a diverse group of women in other geographic areas are needed to evaluate the impact of AI-guided screening on cardiomyopathy diagnosis as well as its impact on adverse maternal outcomes.”
Background and study details:
- A total of 1,195 women between the ages of 18 and 49 were enrolled in the study and completed the required baseline tests. About 73% were pregnant at the beginning of the study; and 39% were in the third trimester.
- Study participants were enrolled in six teaching hospitals in Nigeria.
- The women were randomly assigned, with 587 women in the AI-assisted intervention arm and 608 in the control group.
According to the American Heart Association, heart disease is the No. 1 killer of new mothers in the United States. There’s a maternal health crisis in America, and a 2023 joint statement from the Association of Black Cardiologists, the American College of Cardiology and the American Heart Association called pregnancy “nature’s first cardiac stress test” for a woman — providing a window into future cardiovascular health.
The causes of peripartum cardiomyopathy are still unclear, but with treatment, many affected patients return to normal heart function. For some, however, it can lead to more severe heart failure, so early detection and treatment are important, according to the American Heart Association.
Limitations of the study include that the rate of cardiomyopathy observed among women in the study who were enrolled in teaching hospitals may not reflect the rate in the general obstetric population in Nigeria. In addition, the cut-off value selected for the detection of cardiomyopathy in this study was left ventricular ejection fraction of less than 50%; however, this is different from the original training target (less than 40%) of the AI algorithm adapted for use with the digital stethoscope. The authors observed improved detection of ejection fraction less than 40% compared with less than 50%. AHA guidelines define less than 40% reduced ejection fraction as heart failure.
Co-authors, disclosures and funding sources are listed in the abstract.
Opinions and conclusions of studies presented at American Heart Association scientific meetings are solely those of the study authors and do not necessarily reflect the policy or position of the association. The Association makes no representations or warranties as to their accuracy or reliability. Abstracts presented at the association’s scientific meetings are not peer-reviewed, but they are curated by independent review panels and are evaluated based on the potential to add to the diversity of scientific questions and viewpoints discussed at the meeting. The results are considered preliminary until they are published as a full manuscript in a peer-reviewed scientific journal.
The association primarily receives funds from individuals; foundations and corporations (including the pharmaceutical industry, device manufacturers, and other companies) also make donations and fund specific association programs and events. The association has strict policies to prevent these conditions from influencing the scientific content. Revenues from pharmaceutical and biotech companies, device manufacturers and health insurance providers and the association’s overall financial information are available here.
About the American Heart Association
The American Heart Association is a relentless force for a world of longer, healthier lives. We are dedicated to ensuring equal health in all communities. Through collaboration with numerous organizations and powered by millions of volunteers, we fund innovative research, advocate for public health and share life-saving resources. The Dallas-based organization has been a leading source of health information for nearly a century. Connect with us at heart.org, Facebook, x or by calling 1-800-AHA-USA1.
For media inquiries and AHA expert perspective:
AHA Communications & Media Relations in Dallas: 214-706-1173; [email protected]
Bridgette McNeil: [email protected]
For public inquiries: 1-800-AHA-USA1 (242-8721)
heart.org and stroke.org